Keratins K2 and K10 Are Essential for the Epidermal Integrity of Plantar Skin

Journal of Dermatological Science 81 (2016) 10–16

Contents lists available at ScienceDirect

Journal of Dermatological Science

journal homepage: www.jdsjournal.com

Keratins K2 and K10 are essential for the epidermal integrity of plantar skin

a, b c,d a

Heinz Fischer **, Lutz Langbein , Julia Reichelt , Maria Buchberger ,

a a,

Erwin Tschachler , Leopold Eckhart *

Department of Dermatology, Medical University of Vienna, 1090 Vienna, Austria

Department of Genetics of Skin Carcinogenesis, German Cancer Research Center, Heidelberg, Germany

Institute of Cellular Medicine and North East England Stem Cell Institute, Newcastle University, Newcastle upon Tyne, UK

Divison of Experimental Dermatology and EB House Austria, Department of Dermatology, University Hospital Salzburg, Paracelsus Medical University,

Salzburg, Austria

A R T I C L E I N F O A B S T R A C T

Article history: Background: K1 and K2 are the main type II keratins in the suprabasal epidermis where each of them

Received 20 July 2015

heterodimerizes with the type I keratin K10 to form intermediate ﬁlaments. In regions of the ears, tail,

Received in revised form 7 October 2015

and soles of the mouse, only K2 is co-expressed with K10, suggesting that these keratins sufﬁce to form a

Accepted 8 October 2015

mechanically resilient cytoskeleton.

Objective: To determine the effects of the suppression of both main keratins, K2 and K10, in the suprabasal

Keywords:

plantar epidermis of the mouse.

Keratin / / / /

Methods: Krt2 Krt10 mice were generated by crossing Krt2 and Krt10 mice. Epidermal

Gene deletion

morphology of soles of hind-paws was examined macroscopically and histologically. Immunoﬂuores-

Palmoplantar

Sole cence analysis and quantitative PCR analysis were performed to analyze the expression of keratins in sole

/ /

Mouse model skin of wildtype and Krt2 Krt10 mice. Highly abundant proteins of the sole stratum corneum were

determined by electrophoretic and chromatographic separation and subsequent mass spectrometry.

Results: K2 and K10 are the most prominent suprabasal keratins in normal mouse soles with the

exception of the footpads where K1, K9 and K10 predominate. Mice lacking both K2 and K10 were viable

and developed epidermal acanthosis and hyperkeratosis in inter-footpad epidermis of the soles. The

expression of keratins K1, K9 and K16 was massively increased at the RNA and protein levels in the soles / /

of Krt2 Krt10 mice.

Conclusions: This study demonstrates that the loss of the main cytoskeletal components of plantar

epidermis, i.e. K2 and K10, can be only partly compensated by the upregulation of other keratins. The

/ /

thickening of the epidermis in the soles of Krt2 Krt10 mice may serve as a model for

pathomechanistic aspects of palmoplantar keratoderma.

ã 2015 Japanese Society for Investigative Dermatology. Published by Elsevier Ireland Ltd. All rights

reserved.

1. Introduction important cellular processes, such as tissue growth and stress

responses [1,2]. The speciﬁc expression of distinct keratins during

Keratins are the main cytoskeletal proteins of epithelia in which epithelial cell differentiation and the intermolecular binding

they provide mechanical resilience and serve as scaffolds for afﬁnities lead to the formation of cell type and differentiation

stage-dependent type I and type II keratin heterodimers. K5 and

K14 form the dimers of the basal layer of squamous epithelia

Abbreviations: ESI, electrospray ionization; H&E, hematoxylin and eosin; HPLC, whereas different keratins are expressed in the suprabasal layers.

high-performance liquid chromatography; KO, knockout; MS, mass spectrometry; In the epidermis suprabasal keratinocytes express the type II

PCR, polymerase chain reaction; RT, reverse transcription.

keratins K1, K2, K77 and K80 as well as the type I keratins K9 and

* Corresponding author at: Department of Dermatology, Medical University of

K10 in a highly regulated manner [3,4]. The predominant supra-

Vienna, Lazarettgasse 14, 1090 Vienna, Austria. Fax: +43 1 40400 73584.

** Corresponding author at: Department of Dermatology, Medical University of basal epidermal keratin pair is K1/K10 [3]. At mechanically

Vienna, Lazarettgasse 14, 1090 Vienna, Austria. Fax: +43 1 4081287. stressed human body sites K2/K10 form the most abundant dimer

E-mail addresses: heinz.ﬁ[email protected] (H. Fischer),

in the uppermost epidermal layers besides K1/10, and K1/K9 has [email protected] (L. Eckhart).

http://dx.doi.org/10.1016/j.jdermsci.2015.10.008

H. Fischer et al. / Journal of Dermatological Science 81 (2016) 10–16 11

been proposed to form in the ridges of palmoplantar human 2.3. Histological and immunoﬂuorescence analysis

epidermis [5–7]. High epithelial cell turnover, as during wound

healing or in psoriasis, is associated with the expression of keratins Tissues were ﬁxed in phosphate-buffered 7.5 % formaldehyde

K6 and K16 [8–10]. (SAV LP, Flintsbach, Germany) for 24 h at room temperature and

Mutations of several keratin genes are associated with defects embedded in parafﬁn. Sections of 4 mm were stained with

in epidermal integrity [11–13]. Dominant mutations of genes hematoxylin-eosin or subjected to immunoﬂuorescence analysis

coding for K1 and K10 cause epidermolytic ichthyosis (KRT1, as previously described [22]. For nuclear counterstaining Hoechst

KRT10) and mutations of KRT2 cause superﬁcial epidermolytic 33258 (Sigma–Aldrich, St. Louis, MO) was used. For bright-ﬁeld

ichthyosis, previously known as ichthyosis bullosa of Siemens [13– and epiﬂuorescence microscopy an Olympus AX-70 microscope

15]. These diseases are characterized by protein aggregation, (Olympus, Hamburg, Germany) equipped with a Spot RT3 slider

acanthosis, hyperkeratosis and epidermal fragility that manifests camera (SPOT Imaging Solutions, Sterling Heights, Germany) was

at mechanically challenged body sites [16,17]. Mutations in KRT1, used. Pictures from the skin surface were recorded by a Leica

KRT6, KRT9, KRT10 and KRT16 cause pathological epidermal MZ16A Stereo Microscope (Leica Microsystems, Wetzlar,

thickening on palms and soles, manifesting as different forms of Germany).

palmoplantar keratodermas [15,18].

The main in vivo models for the investigation of normal keratin 2.4. Quantiﬁcation of mRNA by reverse transcription and real-time

functions and for keratinopathies are mouse lines in which keratin PCR

genes have been speciﬁcally mutated [19]. Recently, we have

reported the generation of K2-deﬁcient mice [20]. In the mouse, Extraction of mRNA, reverse transcription and real-time PCRs

K1 and K2 are expressed in a mutually exclusive manner on were performed as previously described [23]. The following primers

0 0

different body sites with K1 being expressed in the interfollicular were used: K1 (Krt1) forward 5 -AGCAAGGCTGAAGCTGAGAC-3 ,

0 0 0

epidermis of hairy skin and K2 in the epidermis of glabrous skin on reverse 5 -GGGCATCTTTGAGTGCTTTC-3 ; K2 (Krt2) forward 5 -

0 0

the soles or skin on the tail and the ears, covered with sparse hair GGAAATCAGCGAGTTGAACC-3 , reverse 5 -ATCTCCACATCCAGGGA-

0 0 0

[21,20]. Focusing on the epidermis of the ears, we have shown that CAG-3 ; K9 (Krt9) forward 5 -CAGTCCCAGTCCTCCTCAAA-3 , reverse

0 0 0

the deletion of Krt2 resulted in aberrant aggregation of K10, 5 -ACTCCATCCAGTTCCAAACG-3 ; K10 (Krt10) forward5 -GACAACT-

0 0 0

epidermal hyperkeratosis and skin inﬂammation [20]. The ablation GACAATGCCAACG-3 ,reverse 5 -GCCAGCTCTTCGTTCAGACT-3 ;K14

0 0 0

of Krt2 was partially compensated by the induction of K1 in the (Krt14) forward 5 -ATCGAGGACCTGAAGAGCAA-3 , reverse 5 -

0 0

epidermis of the ear. The concomitant deletion of both K2 and GGCTCTCAATCTGCATCTCC-3 ; K16 (Krt16) forward 5 -AGAACCG-

0 0 0

K10 suppressed aberrant keratin aggregation but also greatly CAGAGATGTGGAG-3 , reverse 5 -CTGCTGATCAAACCCTGGAT-3 .

diminished the intermediate ﬁlament network and resulted in

hyperkeratosis. The skin on the soles of K2-deﬁcient and K2/K10- 2.5. Preparation of stratum corneum extracts

deﬁcient mice has remained uncharacterized.

Here we report the sole phenotype of K2/K10 double gene Stratum corneum was scraped off from plantar areas of the

knockout mice. Our results show the ability of plantar keratinocytes hind-sole of euthanized mice using a scalpel. Special attention was

to activate a compensatoryupregulation ofkeratins K1, K9and K16 in paid to avoid contamination with footpad material. For extraction

the absence of K2 and K10. These ﬁndings shed new light on the of proteins, the samples were transferred into lysis buffer

normal function of K2 and K10 in the epidermis of the soles and containing 2% sodium dodecyl sulfate (SDS) (Sigma–Aldrich, St.

suggest that palmoplantar keratin gene expression can adapt to the Louis, MO), 100 mM dithiothreitol (Sigma–Aldrich, St. Louis, MO)

suppression of these most abundant keratins, however, with limited and 50 mM Tris–HCl, pH 7.6, sonicated for 20 s, and incubated at

potential for compensation for their function. 95 C for 10 min. Samples were cleared by repeated centrifugation

and supernatants were stored at 20 C.

2. Materials and methods

2.6. Protein electrophoresis

2.1. Mice

Stratum corneum extracts were electrophoresed through an

Krt2 knockout mice [20] on a C57BL/6 background and Krt10 8–18% gradient polyacrylamide gel, and polypeptides were

knockout mice on a Balb/c background [7] were crossed to stained using EZBlue gel staining reagent (Sigma–Aldrich, St.

/ /

generate Krt2 Krt10 mice. Mice were maintained according to Louis, MO) according to the manufacturer’s instructions. Protein

the animal welfare guidelines of the Medical University of Vienna, bands of interest were excised using clean and sterile scalpels in a

Austria. Genotyping was done as reported [20]. laminar ﬂow hood. Gel pieces were stored in water until further analysis.

2.2. Antisera

2.7. Liquid chromatography, tandem mass spectrometry (LC-MS/MS)

Antisera for immunoﬂuorescence were as follows: anti-K2 and protein identiﬁcation

(1:2000), anti-K5 (1:5000) and anti-K9 (1:2000) were raised in

guinea pigs, as previously described (raised in guinea pigs by L.L., Processing of samples and LC-MS/MS were performed by

available from Progen, Heidelberg, Germany). Rabbit anti-K1 VetCORE, University of Veterinary Medicine Vienna (Vienna,

(1:1000), rabbit anti-K6 (1:500) and rabbit anti-K10 (1:1000) Austria). Brieﬂy, the samples were reduced with DTT, alkylated

were purchased from Covance (Fremont, CA), and rabbit anti-K16 with iodoacetamide, and digested with trypsin (Trypsin Gold, mass

(1:1000) were obtained from Acris (Herford, Germany). The spectrometry grade, Promega, Madison, WI) in the gel. Peptides

ﬂuorescence-labelled secondary antibodies, all raised in goat, were separated by reversed-phase HPLC on a 50 cm Acclaim

were anti-guinea pig Alexa Fluor 594 (1:500), anti-rabbit Alexa PepMap C18 column. For mass spectrometric analysis, an HCT

Fluor 546 (1:500) (Invitrogen, Carlsbad, CA). Esquire mass spectrometer (Bruker Daltonics, Bremen, Germany)

was used in auto-MS(n) run mode. Automatic MS/MS was

performed in peptide scan mode with the following settings:

MS(n) averages: 2; precursor ions: 3; active exclusion on (after

12 H. Fischer et al. / Journal of Dermatological Science 81 (2016) 10–16

1 spectra for 0.2 min); exclude single charged ions; MS/MS scan two different modes of keratinization in the mouse sole: One

range: 100–1800. involving the formation of intermediate ﬁlaments made of K2/

Processed spectra were searched via a Mascot server version 2.3 K10 heterodimers in the regions of plantar skin outside of the

(Matrix Science, Boston, MA) and the software ProteinScape 2.1 footpads and the other involving the formation of intermediate

(Bruker Daltonics, Bremen, Germany) in the UniProt database (Mus ﬁlaments composed of K1/K9 and K1/K10 heterodimers in the

musculus only) using the following search parameters: global footpads.

modiﬁcation carbamidomethylation on cysteine; variable mod-

iﬁcations oxidation on methionine; deamidation on asparagine 3.2. Deletion of Krt2 and Krt10 causes acanthosis, hyperkeratosis and

and glutamine as well as formation of pyroglutamic acid; enzyme scaling of the stratum corneum in plantar skin

speciﬁcity trypsin; charge state z = 2+, 3+; MS tolerance 0.4 Da; MS/

MS tolerance 0.4 Da; one missed cleavage allowed; signiﬁcance To investigate of roles of the dimerizing keratins K2 and K10 in

threshold p < 0.05. vivo, the genes encoding both keratins were deleted: animals

carrying Krt2 [20] and Krt10 [7] single gene knockouts were

3. Results crossed as described previously [20]. Absence of K2 and

K10 expression in skin was conﬁrmed on the RNA and protein

3.1. Keratins K1, K2, and K9 are differently expressed in the sole levels by quantitative RT-PCR (Supplementary Fig. S1B) and

epidermis of the mouse immunoﬂuorescence analysis (Supplementary Fig. S1A), respec-

tively. Single gene knockout, heterozygous, and double gene

The sole of mice is organized in the footpads and inter-footpad knockout mice were born at Mendelian ratios, and showed no

regions, as depicted schematically in Fig. 1A. The expression signiﬁcant differences in size and body weight [20].

patterns of the main suprabasal keratins, K1, K2, K9, and K10 were The deletion of K2 and K10 resulted in massive aberrations of

determined by immunoﬂuorescence labeling of the skin derived the inter-footpad regions of plantar skin whereas the footpads

from mouse soles (Fig. 1B). K10 was detected in the suprabasal themselves appeared widely normal (Figs. 2 and 3; Supplementary

layers of the epidermis throughout the entire sole. By contrast, Fig. S2). Macroscopically, the surface of plantar skin of Krt2

K2 was prominently expressed in the plantar epidermis outside of Krt10 mice showed hyperkeratosis and ﬂaking (Fig. 2). Besides

/ /

the footpads and in sparse cells of the footpad whereas K1 and the soles, the toes of Krt2 Krt10 mice were affected in a

K9 were expressed in the footpads only. The mutually exclusive similar manner (Fig. 2, middle panels). Hematoxylin and eosin

expression pattern of K2 versus K1 and K9 suggests the existence of (H&E)-staining revealed prominent acanthosis and hyperkeratosis

Fig. 1. Suprabasal epidermal keratins are differentially expressed in plantar skin.

(A) Schematic depiction of the hind-paw of a mouse. The glabrous skin contains Fig. 2. Deletion of Krt2,Krt10 or both causes acanthosis, hyperkeratosis and scaly

footpads (arrows). The site of tissue sampling for sole skin outside of footpads is stratum corneum in plantar skin. Representative photos of the hind-paws obtained

/ / / /

indicated by an asterisk. (B) Immunoﬂuorescence labeling of K1, K2, K9 and K10 in from wildtype, Krt2 , Krt10 , and Krt2 Krt10 mice (upper panels). Details

sole skin (marked by an asterisk in panel A) outside of footpads and in the footpads. (marked by frames in the uppermost panels) of toes and soles are shown at higher

Dotted and broken lines mark the border of the stratum corneum and the dermo- magniﬁcation in the middle and lowermost panels, respectively. Open arrows point

epidermal junction, respectively. Scale bars: 50 mm. to calluses and corniﬁed scales present in regions outside of the footpads (arrows).

H. Fischer et al. / Journal of Dermatological Science 81 (2016) 10–16 13

(Supplementary Fig. S3). Taken together these data showed that

the concomitant deletion of both suprabasal keratins, i.e. K2 and

K10, causes a more severe phenotype in the mouse inter-footpad

epidermis than the deletion of either K2 or K10 alone, and suggest

that K2 and K10 play a synergistic role for the establishment of a

normal epidermal barrier on plantar skin.

3.3. Deletion of Krt2 and Krt10 causes upregulation of K1, K9 and

K16 in sole skin

To investigate consequences of the K2/K10 deletion on the

protein composition of plantar stratum corneum, we extracted

proteins from wildtype and mutant mouse soles excluding

footpads and separated them by SDS polyacrylamide gel electro-

phoresis (PAGE) (Fig. 4). In samples of wildtype mice, a staining of

two major protein bands at the predicted molecular weights of

K2 and K10 were readily detectable. These bands were absent in

extracts of K2 and K10 knockout mice, respectively, demonstrating

that these keratins are the predominant protein species in stratum

corneum extracts of murine plantar epidermis. In the stratum

/ /

corneum extracts from Krt2 Krt10 mice, both K2 and

K10 bands disappeared, as expected, whereas several other protein

bands appeared (Fig. 4), indicating an altered protein composition.

To identify these accumulating proteins in the sole stratum

/ /

corneum of Krt2 Krt10 mice, 3 segments of the gel

corresponding to either single or double-bands were excised

and subjected to mass spectrometric analysis using nano-HPLC

ESI–MS/MS. In the gel segments corresponding to proteins of

around 67–73 kDa (P1), 57–63 kDa (P2) and 43–48 kDa (P3) the

keratins K9 and K1 (P1), K1 (P2) and K16 (P3) were most abundant,

respectively (Fig. 4, for the complete list of proteins identiﬁed see

Supplementary Fig. S4).

The expression of these keratins was further analyzed at the

Fig. 3. Epidermal hyperplasia of plantar skin is further increased upon genetic

protein and mRNA level (Fig. 5A and B). Immunoﬂuorescence

deletion of both Krt2 and Krt10. (A) Hematoxylin and eosin (H&E) staining of plantar / /

labeling of Krt2 Krt10 mice conﬁrmed signiﬁcant upregula-

skin. Knockout of Krt2, Krt10 or both results in several prominent alterations within

tion of K1 and K16 in the suprabasal layers of inter-footpad

the epidermis, including acanthosis (diamonds), hypergranulation (arrows) and

hyperkeratosis (asterisks). Scale bars: 50 mm. (B) Quantiﬁcation of the thickness of epidermis as compared to wildtype samples (Fig. 5A). Double-

plantar epidermis. The mean thickness of the living epidermis (indicated by a

labeling experiments revealed nearly perfect co-localization of K1

bracket) was measured using Metamorph image analysis software over the length

and K16 (Supplementary Fig. S5). K6 was expressed more basally

of one skin section per mouse. At least 3 mice per genotype were investigated.

than K16, and double-labeling of K1 and K6 showed little co-

Individual values and the mean are shown. Error bars show the standard errors of

the mean. Statistical signiﬁcance was calculated using analysis of variance (ANOVA) localization, indicating that K16 preferentially formed hetero-

with Tukeys multiple comparisons test. Differences with p-values smaller than dimers with K1 but not, or only to a minor extent, with its

0.01 are marked with an asterisk.

of the plantar epidermis outside the footpads (Fig. 3A). The living

part of the epidermis was approximately 2.5-fold thicker in Krt2

/Krt10 than in wildtype mice (p < 0.0001) and also signiﬁcantly

/ /

thicker than in Krt2 mice and Krt10 mice (Fig. 3B). In the

/ /

granular layer of Krt2 /Krt10 mice, accumulation and massive

enlargement of keratohyalin granules with occasional cytolysis

and blister formation was detected (Supplementary Fig. S3).

Hyperkeratosis, keratohyalin accumulation and cytolysis and

blistering were also present in the plantar epidermis of Krt2

and Krt10 mice, however, the morphological aberrations and

epidermal defects of single gene knockout mice appeared to be

milder than those in the double knockout mice (Supplementary

Fig. S3). The irregular appearance of cytolysis and intra-epidermal

Fig. 4. Deletion of K2 and K10 causes a fundamental change of the polypeptide

blistering was most likely caused by stress but remains to be

pattern extracted from plantar stratum corneum. Proteins were extracted from the

investigated further.

/ / / stratum corneum of the extra-footpad region of sole skin (Fig. 1, asterisk), separated

The stratum corneum of the soles of Krt2 , Krt10 and Krt2

by SDS-PAGE and stained with EZBlue. Arrows point to bands corresponding to the

Krt10 mice was thickened and less compact than that of sizes of K2 and K10 that could be found in extracts of wildtype mice but not in the

wildtype mice. The disturbance of the stratum corneum morphol- respective knockout mouse samples. Major bands detected in extracts from Krt2

/ / Krt10 mice (P1–P3) were cut out and subjected to nano-HPLC ESI–MS/MS. Most

ogy was most severe in Krt2 Krt10 mice where the stratum

prominent polypeptides identiﬁed (MS score >1000 in P1 and P2; MS score >5000 in

corneum had an extremely spongy or basked-waved appearance

P3) are listed. Information about further protein hits is provided in Supplementary

and showed aberrantly high staining afﬁnity for hematoxylin Fig. S4.

14 H. Fischer et al. / Journal of Dermatological Science 81 (2016) 10–16

Quantitative PCR conﬁrmed strong upregulation of suprabasal

keratins K1, K9 and K16, whereas the expression level of basal

keratin K14 was not altered in the inter-footpad skin of Krt2

Krt10 mice (Fig. 5B). In summary, the absence of K2 and

K10 results in compensatory upregulation of other keratin species,

including co-expression of K1 and K16, indicating that they

substitute for the absence of the K2/K10 intermediate ﬁlaments.

4. Discussion

This study deﬁnes an essential role of K2/K10 intermediate

ﬁlaments in plantar skin of mice. The results demonstrate (1) that

K2 and K10 are the 2 most abundant protein species in the stratum

corneum of the mouse sole outside of the footpad and (2) that

concomitant absence of both keratins is incompatible with normal

keratinocyte differentiation. The targeted deletion of K2 and

K10 establishes a mouse model in which abnormalities of the

epidermis on the soles are the most severe phenotype under the

standard housing conditions of mice, and therefore may be useful

in the investigation of human diseases that manifest in structural

defects of palmoplantar epidermis.

The characterization of the sole skin extends our previous study

of mice carrying a deletion of K2 [20]. In line with the restricted

expression of K2 in wildtype mice, the phenotype of Krt2 and

/ /

Krt2 Krt10 the gene knockout mice is most severe on the ears

[20], the interscale regions [24] of the tail (Supplementary Fig. S6)

and the soles whereas deletion of K2 does not affect hairy skin.

Lack of normal K2 expression leads to acanthosis and hyperkera-

tosis as well as to increased abundance of keratohyalin granules in

the suprabasal epidermis of ears, tail and the sole.

The results of this study show that the suppression of K2 and

K10 leads to a redistribution of K14 immunoreactivity from

exclusively basal to all epidermal layers and to the transcriptional

upregulation of K1, K9, and K16 in the suprabasal plantar

epidermis. Our data on the colocalisation of K1 and K16 and the

lack of colocalization of the expected dimerization partners K6 and

K16 [19] in the suprabasal plantar epidermis indicate that K1 and

/ /

K16 might heterodimerize in the soles of Krt2 Krt10 mice. In

line with the concept of an interdependency or interaction of

K1 and K16, K1 appeared to undergo aggregation in calluses on the

paws of Krt16 mice [25]. Other keratins such as K77 [4], K80 [3],

and keratins detected at lower abundance by ESI–MS/MS analysis

/ /

of Krt2 Krt10 sole stratum corneum (Fig. 4; Supplementary

Fig. S4) may also contribute to the homeostasis of plantar

epidermis. Together with previous reports [7,9,26–28], our data

suggest that the expression of keratins is subjected to complex

regulation and allows for a considerable plasticity in keratin

Fig. 5. Deletion of K2 and K10 results in upregulation of K1, K9 and K16. (A)

/ / heterodimerization patterns in the epidermis of the sole.

Immunoﬂuorescence analysis of inter-footpad skin derived from of Krt2 Krt10

ﬁ

and wild-type mice using antisera against K1, K9, K14 and K16 (red). Nuclei were Although the amounts of K1, K9 and K16 appear to be signi cant

/ /

stained by using Hoechst 33258 (blue). Dotted and broken lines mark the borders of in Krt2 Krt10 stratum corneum (as suggested by the protein

the stratum corneum and the dermo-epidermal junction, respectively. Scale bars:

pattern of electrophoresed extracts), their presence fails to

40 mm. (B) mRNA obtained from inter-footpad skin of at least 3 mice per genotype

compensate for the lack of K2 and K10 in establishing a normal

was analyzed by quantitative PCR for the expression of K1, K9, K14 and K16.

epidermal stratiﬁcation and stratum corneum formation. The

Expression levels of the keratins (arbitrary units, a.u.) were calculated relative to

+/+ /

that of the house-keeping gene beta-2 microglobulin. Error bars show the standard ﬁnding that the plantar skin phenotypes of Krt2 Krt10 and

/ +/+ / /

error of the mean. *, p < 0.05; ns, not signiﬁcant (t-test). (For interpretation of the

Krt2 Krt10 mice were less severe than that of Krt2 Krt10

references to color in this ﬁgure legend, the reader is referred to the web version of

mice suggests that K2 and K10 individually can contribute, to some

this article.)

extent, to keratinocyte differentiation even when their respective

optimal binding partner is missing.

“ ”

expected binding partner K6 (Supplementary Fig. S5). Although Notably, the phenotype of K2 knockout mice [20] is generally

/ /

K9 was found to be upregulated in Krt2 Krt10 mice its pattern less severe than that of K1-deﬁcient mice [29] and the phenotype

of expression was patchy and limited to a small number of of K2/K10 double knockout mice [20; present study] is less severe

suprabasal cells (Fig. 5A). Immunolabeling of K14 indicated a shift that of K1/K10 double knockout mice [30], which die perinatally.

of from a strictly basal pattern in wildtype mice to the presence of The differences are likely attributable to the fact that the areas of

/ /

K14 in basal and suprabasal epidermal layers in Krt2 Krt10 K2 expression on the ears, tail, and soles are much smaller than the

mice (Fig. 5A). area of K1 expression in the epidermis of essentially all other body

sites [20].

H. Fischer et al. / Journal of Dermatological Science 81 (2016) 10–16 15

The deletion of K10 disturbs the inter-footpad sole but does not References

affect epidermal integrity of footpads. This suggests that K9 alone

[1] P.A. Coulombe, P. Wong, Cytoplasmic intermediate ﬁlaments revealed as

sufﬁces to heterodimerize and form ﬁlaments with K1 at this

dynamic and multipurpose scaffolds, Nat. Cell Biol. 6 (2004) 699–706.

particular skin site. This hypothesis is supported by the epider-

[2] L. Ramms, G. Fabris, R. Windoffer, N. Schwarz, R. Springer, C. Zhou, J. Lazar, S.

molytic palmoplantar keratoderma-like phenotype of the footpads Stiefel, N. Hersch, U. Schnakenberg, T.M. Magin, R.E. Leube, R. Merkel, B.

Hoffmann, Keratins as the main component for the mechanical integrity of

of K9-deleted mice [31]. Although K9 null mice exhibit no major

keratinocytes, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 18513–18518.

changes in the level and distribution of K1 and K10, the structural

[3] L. Langbein, L. Eckhart, M.A. Rogers, S. Praetzel-Wunder, J. Schweizer, Against

integrity of the footpad epidermis was severely affected indicating the rules: the human keratin K80-two functional alternative splice variants,

K80 and K80.1, with special cellular localization in a wide range of epithelia, J.

that K1/K10 cannot compensate the loss of K1/K9 heterodimers.

/ Biol. Chem. 285 (2010) 36909–36921.

The structural disturbance of the footpad skin in Krt9 mice is

[4] L. Langbein, J. Reichelt, L. Eckhart, S. Praetzel-Wunder, W. Kittstein, N. Gassler,

/ /

less severe than that of Krt2 Krt10 mice, although the J. Schweizer, New facets of keratin K77: interspecies variations of expression

footpads (sites of K9 expression) are likely exposed to higher loads and different intracellular location in embryonic and adult skin of man and

mouse, Cell Tissue Res. 354 (2013) 793–812.

of mechanical stress than the inter-footpad regions (sites of

[5] L. Langbein, H.W. Heid, I. Moll, W.W. Franke, Molecular characterization of the

K2 expression). It remains to be investigated whether external

body site-speciﬁc human cytokeratin 9: cDNA cloning, amino acid sequence

ﬁ –

stress or intrinsic disorganization of cellular structures are the and tissue speci city of gene expression, Differentiation 55 (1993) 57 73.

[6] O. Swensson, L. Langbein, J.R. McMillan, H.P. Stevens, I.M. Leigh, W.H. McLean,

main drivers of sole skin defects in these mouse models.

E.B. Lane, R.A. Eady, Specialized keratin expression pattern in human ridged

The thickening of the plantar epidermis and the loosening of the

skin as an adaptation to high physical stress, Br. J. Dermatol. 139 (1998) 767–

/ /

layers of the stratum corneum in Krt2 Krt10 mice shares 775.

[7] J. Reichelt, H. Büssow, C. Grund, T.M. Magin, Formation of a normal epidermis

some but not all histological aspects with certain palmoplantar

supported by increased stability of keratins 5 and 14 in keratin 10 null mice,

keratodermas in human patients [11,15,18]. In humans palmo-

Mol. Biol. Cell 12 (2001) 1557–1568.

plantar keratoderma is caused by mutations in structurally diverse [8] R.D. Paladini, K. Takahashi, N.S. Bravo, et al., Onset of re-epithelialization after

skin injury correlates with a reorganization of keratin ﬁlaments in wound edge

proteins including keratins K1, K6, K9, K10 and K16, corniﬁed

keratinocytes: deﬁning a potential role for keratin 16, J. Cell Biol. 132 (1996)

envelope proteins, connexins, desmosomal proteins and cathe- 381–397.

psins [32,33], indicating that various defects in the molecular [9] J.C. Lessard, S. Piña-Paz, J.D. Rotty, R.P. Hickerson, R.L. Kaspar, A. Balmain, P.A.

Coulombe, Keratin 16 regulates innate immunity in response to epidermal

architecture of plantar epidermis converge into common (or at

barrier breach, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 19537–19542.

least similar) clinical phenotypes. Although the rare pathogenic

[10] C. Hivnor, N. Williams, F. Singh, A. VanVoorhees, L. Dzubow, D. Baldwin, J.

K2 mutations have not been reported to cause palmoplantar Seykora, Gene expression proﬁling of porokeratosis demonstrates similarities

–

keratoderma [6,34,35], and the concomitant inactivation of with psoriasis, J. Cutan. Pathol. 31 (2004) 657 664.

[11] M.B. Omary, P.A. Coulombe, W.H. McLean, Intermediate ﬁlament proteins and

2 keratins does not replicate any known human pathology at

–

/ / their associated diseases, N. Engl. J. Med. 351 (2004) 2087 2100.

the molecular level, the Krt2 Krt10 mice represent an

[12] J. Uitto, G. Richard, J.A. McGrath, Diseases of epidermal keratins and their

–

interesting model for investigating keratin functions that are linker proteins, Exp. Cell Res. 313 (2007) 1995 2009.

[13] W.H. McLean, C.B. Moore, Keratin disorders: from gene to therapy, Hum. Mol.

relevant for skin diseases.

Genet. 20 (R2) (2011) R189–R197.

The compensatory upregulation of keratins in situations of keratin

[14] W.H.I. McLean, S.M. Morley, E.B. Lane, R.A.J. Eady, W.A.D. Grifﬁths, D.G. Paige, J.

mutations appears to be a natural response of sole keratinocytes that I. Harper, C. Higgins, I.M. Leigh, Ichthyosis bullosa of Siemens–a disease

involving keratin 2e, J. Invest. Dermatol. 103 (1994) 277–281.

might be utilized therapeutically in hereditary keratinopathies, as it

/ / [15] M. Knöbel, E.A. O'Toole, F.J. Smith, Keratins and skin disease, Cell Tissue Res.

has been suggested previously [36]. The Krt2 Krt10 mouse

360 (2015) 583–589.

ﬁ

model may help to de ne the signaling pathways that trigger the [16] J.C. Chamcheu, I.A. Siddiqui, D.N. Syed, V.M. Adhami, M. Liovic, H. Mukhtar,

Keratin gene mutations in disorders of human skin and its appendages, Arch.

upregulation of speciﬁc keratins in the sole epidermis. Intriguingly,

Biochem. Biophys. 508 (2011) 123–137.

the development of intra-epidermal blistering is associated with

[17] J.C. Chamcheu, G.S. Wood, I.A. Siddiqui, D.N. Syed, V.M. Adhami, J.M. Teng, H.

/ /

ﬂ

little if any in ammation in the soles of Krt2 Krt10 mice Mukhtar, Progress towards genetic and pharmacological therapies for keratin

genodermatoses: current perspective and future promise, Exp. Dermatol. 21 whereasit resultsina prominent inﬂammatory responseon the ears

(2012) 481–489.

of the same animals. Thus, further studies are required to deﬁne the

[18] J.A. Rothnagel, A.M. Dominey, L.D. Dempsey, M.A. Longley, D.A. Greenhalgh, T.

signaling pathways that underlie changes of keratin gene expres- A. Gagne, M. Huber, E. Frenk, D. Hohl, D.R. Roop, Mutations in the rod domains

–

sion and control the homeostasis of plantar skin. of keratins 1 and 10 in epidermolytic hyperkeratosis, Science 257 (1992) 1128

1130.

[19] P.A. Coulombe, X. Tong, S. Mazzalupo, Z. Wang, P. Wong, Great promises yet to

Conﬂict of interest

be fulfilled: defining keratin intermediate filament function in vivo, Eur. J. Cell

Biol. 83 (2004) 735–746.

[20] H. Fischer, L. Langbein, J. Reichelt, S. Praetzel-Wunder, M. Buchberger, M.

The authors have no conﬂict of interest to declare.

Ghannadan, E. Tschachler, L. Eckhart, Loss of keratin K2 expression causes

aberrant aggregation of K10, hyperkeratosis and inﬂammation, J. Invest.

Acknowledgements Dermatol. 134 (2014) 2579–2588.

[21] M. Rentrop, R. Nischt, B. Knapp, J. Schweizer, H. Winter, An unusual type-II 70-

kilodalton keratin protein of mouse epidermis exhibiting postnatal body-site

Mice carrying a hemizygous deletion of Krt2 were obtained

speciﬁcity and sensitivity to hyperproliferation, Differentiation 34 (1987) 189–

from the International Knockout Mouse Consortium (KOMP/ 200.

[22] H. Fischer, L. Eckhart, M. Mildner, K. Jaeger, M. Buchberger, M. Ghannadan, E.

EUCOMM). E.T. and L.E. were supported by COST Action TD1206

Tschachler, DNase1L2 degrades nuclear DNA during corneocyte formation, J.

StanDerm. This study was also supported in part by the Wilhelm

Invest. Dermatol. 127 (2007) 24–30.

Sander-Stiftung, Munich, Germany (Grant 2007.133.2 to L.L.) and [23] H. Fischer, S. Szabo, J. Scherz, K. Jaeger, H. Rossiter, M. Buchberger, M.

Ghannadan, M. Hermann, H.C. Theussl, D.J. Tobin, E.F. Wagner, E. Tschachler, L.

by Newcastle Health Care Charity and the Newcastle upon Tyne

Eckhart, Essential role of the keratinocyte-speciﬁc endonuclease DNase1L2 in

Hospitals NHS Charity (Grant PFC/ ML/0809 to J.R.). We thank

the removal of nuclear DNA from hair and nails, J. Invest. Dermatol. 131 (2011)

Harald Hoeger for maintaining the mice and Karin Sautner 1208–1215.

(Medical University of Vienna) for technical support. [24] C. Gomez, W. Chua, A. Miremadi, S. Quist, D.J. Headon, F.M. Watt, The

interfollicular epidermis of adult mouse tail comprises two distinct cell

lineages that are differentially regulated by Wnt, Edaradd, and Lrig1, Stem Cell

Appendix A. Supplementary data

Rep. 1 (2013) 19–27.

[25] J.C. Lessard, P.A. Coulombe, Keratin 16-null mice develop palmoplantar

keratoderma, a hallmark feature of pachyonychia congenita and related

Supplementary data associated with this article can be

disorders, J. Invest. Dermatol. 132 (2012) 1384–1391.

found, in the online version, at http://dx.doi.org/10.1016/j. jdermsci.2015.10.008.

16 H. Fischer et al. / Journal of Dermatological Science 81 (2016) 10–16

[26] M. Wawersik, R.D. Paladini, E. Noensie, P.A. Coulombe, A proline residue in the [31] D.J. Fu, C. Thomson, D.P. Lunny, P.J. Dopping-Hepenstal, J.A. McGrath, F.J. Smith,

alpha-helical rod domain of type I keratin 16 destabilizes keratin W.H. McLean, D.M. Pedrioli, Keratin 9 is required for the structural integrity

heterotetramers, J. Biol. Chem. 272 (1997) 32557–32565. and terminal differentiation of the palmoplantar epidermis, J. Invest.

[27] F.B. Müller, M. Huber, T. Kinaciyan, I. Hausser, C. Schaffrath, T. Krieg, D. Hohl, B. Dermatol. 134 (2014) 754–763.

P. Korge, M.J. Arin, A human keratin 10 knockout causes recessive [32] R.C. Ratnavel, W.A. Grifﬁths, The inherited palmoplantar keratodermas, Br. J.

epidermolytic hyperkeratosis, Hum. Mol. Genet. 15 (2006) 1133–1141. Dermatol. 137 (1997) 485–490.

[28] Y.A. Cao, R.P. Hickerson, B.L. Seegmiller, D. Grapov, M.M. Gross, M.R. Bessette, [33] A. Kimyai-Asadi, L.B. Kotcher, M.H. Jih, The molecular basis of hereditary

B.S. Phinney, M.A. Flores, T.J. Speaker, A. Vermeulen, A.A. Bravo, palmoplantar keratodermas, J. Am. Acad. Dermatol. 47 (2002) 327–346.

A.L. Bruckner, L.M. Milstone, M.E. Schwartz, R.H. Rice, R.L. Kaspar, Gene [34] J.A. Rothnagel, H. Traupe, S. Wojcik, M. Huber, D. Hohl, M.R. Pittelkow, H. Saeki,

expression proﬁling in pachyonychia congenita skin, J. Dermatol. Sci. 77 (2015) Y. Ishibashi, D.R. Roop, Mutations in the rod domain of keratin 2e in patients

156–165. with ichthyosis bullosa of Siemens, Nat. Genet. 7 (1994) 485–490.

[29] W. Roth, V. Kumar, H.D. Beer, M. Richter, C. Wohlenberg, U. Reuter, S. Thiering, [35] E. Akasaka, S. Minakawa, D. Rokunohe, Y. Toyomaki, Y. Matsuzaki, D.

A. Staratschek-Jox, A. Hofmann, F. Kreusch, J.L. Schultze, T. Vogl, J. Roth, J. Sawamura, H. Nakano, Superﬁcial epidermolytic ichthyosis caused by a novel

Reichelt, I. Hausser, T.M. Magin, Keratin 1 maintains skin integrity and KRT2 mutation, J. Dermatol. Sci. 79 (2015) 86–88.

participates in an inﬂammatory network in skin through interleukin-18, J. Cell. [36] M. Kerns, D. DePianto, M. Yamamoto, P.A. Coulombe, Differential modulation

Sci. 125 (2012) 5269–5279. of keratin expression by sulforaphane occurs via Nrf2-dependent and

[30] L. Wallace, L. Roberts-Thompson, J. Reichelt, Deletion of K1/K10 does not -independent pathways in skin epithelia, Mol. Biol. Cell 21 (2010) 4068–4075.

impair epidermal stratiﬁcation but affects desmosomal structure and nuclear

integrity, J. Cell Sci. 125 (2012) 1750–1758.